Jump to content
Second Skin Audio

Port compression and optimization


Recommended Posts

I was reading an article put out by some JBL engineers (Salvatti, Devantier, & Button) about port optimization. They were doing some really interesting tests with port compression and how much it can hurt output. They were showing port compression causing losses of almost 10 db, which is obviously a huge loss in output. The paper left me with some questions that I'd like to get answers to, so I've decided to do my own testing.
Here are the questions I'm going to try to get answers to:
  1. How does port velocity relate to port compression?
  2. How does port noise relate to port compression?
  3. Do larger diameter ports tolerate higher velocities before an equal level of port compression takes place?
  4. Do larger diameter ports tolerate higher velocities before port noise becomes noticeable?
  5. What other methods can be used to determine f port compression is happening?
The test setup:
To conduct the tests I built a 4 cu ft net box. I used OSB since its really cheap and it doesn’t matter how the box looks. The box has interchangeable external ports to make changing port configuration easy. Here is what it looks like:
0BWVRyK.jpg
The ports I have built for it now are dual 3" ports, a 4" port, and a 5" port.
Other equipment I'll be using for the test is:
- Alpine SWR-15D2 subwoofer
- Behringer NU3000DSP amplifier
- My old HP shop laptop
- Behring UCA-202 USB DAC
- Dayton UMM-6 mic
- SMD AMM-1
- My calibrated super-scientific ears
I got everything put together so I started doing a little testing in the shop tonight. For this round I just verified the tuning frequency of each port and listened for what output level I can first hear port noise at.
The results:
Dual 3" - 33.4 Hz tuning - port noise at 14.6 volts - 53 watts - 18.5 m/s
4" - 33.5 Hz tuning - port noise at 12.8 volts - 41 watts - 18.2 m/s
5" - 34.5 Hz tuning - port noise at 18.1 volts - 82 watts - 17.8 m/s
My tunings all come out pretty close, with the 5" port coming in at just 1 hz higher than the other two. I was shooting for a 35 hz tune so they all are a little low from that, but that's not unusual. Unfortunately I don't have any way to actually measure port velocity, all I can do is plug the numbers into WinISD and see what it tells me the port velocity should be. Here is a graph of all three ports:
4kyupnk.png
The green line is the dual 3" port, the yellow line (which you can't hardly see under the green line) is the 4" port, and the red line is the 5" port.
Interestingly enough all three ports had very similar velocities where I could first hear port noise. The port velocity where this took place was lower than I expected. Though I had my ears about a foot from the ports and I was listening for port noise as hard as I could. It would take quite a bit more port noise before it would really be noticeable in a normal listening environment.
Next steps:
So what's next is the real testing, which I'm going to do outside so the acoustic hell-hole that is my woodshop doesn't taint the results.
I'm going to play sine wave sweeps from 25 - 150 hz at different power levels and measure the output. I can then adjust the graphs so they line up at a higher frequency were the output is coming from the cone and not the port. The difference in the output shown in the graphs around tuning with the higher power sweeps vs the low power sweeps will have been caused by port compression and I should be able to measure it. This method for measuring port compression was not my idea. It came from the paper I was reading. I think it a pretty novel way to test and I'm excited to see how it turns out. It may be a week or so before I get a chance to do the outside testing, but I'll be sure to post my results in this thread when I'm done.

"Nothing prevents people from knowing the truth more than the belief they already know it."
"Making bass is easy, making music is the hard part."

Builds:

U7qkMTL.jpg  LgPgE9w.jpg  Od2G3u1.jpg  xMyLoO1.jpg  9pAlXUK.jpg

Link to comment
Share on other sites

  • Replies 40
  • Created
  • Last Reply
Link to comment
Share on other sites

very interesting indeed, ive read some papers published by jbl, seemed very knowledgeable and interesting. they were testing bunches of things to do with ports and how they react. im always in for some science

Fidelity

Built to Last

Team Sound Asleep

24Runner Build Log: http://www.stevemeadedesigns.com/board/topic/196657-24runner-sleeper-system-lots-of-fi-neo-dd-focal-new-video-w-juicebox-lithium/

2 x 12" Fi BTL N2 / 2 x 12" Fi BTL N3

2 DD M3b

Maxwell 2.7V 3000F Supercapacitors

Pioneer DEH-80prs

Focal P165 V30 components

Rockford Fosgate T-400.4

Link to comment
Share on other sites

very interesting indeed, ive read some papers published by jbl, seemed very knowledgeable and interesting. they were testing bunches of things to do with ports and how they react. im always in for some science

Sounds like the same paper I was reading, they tested out a bunch of different stuff. The paper I read was called "Maximizing Performance from Loudspeaker Ports" It should turn up in Google.

"Nothing prevents people from knowing the truth more than the belief they already know it."
"Making bass is easy, making music is the hard part."

Builds:

U7qkMTL.jpg  LgPgE9w.jpg  Od2G3u1.jpg  xMyLoO1.jpg  9pAlXUK.jpg

Link to comment
Share on other sites

I love all the tests you do...it seems to be everything I want to try but never can make the time for.

I need to make it out your way one of these days and we can do some experimenting with 6ths! Haha

Sure thing man, any time!

6th order designs are something that still have me perplexed and I'd like to experiment with them more.

"Nothing prevents people from knowing the truth more than the belief they already know it."
"Making bass is easy, making music is the hard part."

Builds:

U7qkMTL.jpg  LgPgE9w.jpg  Od2G3u1.jpg  xMyLoO1.jpg  9pAlXUK.jpg

Link to comment
Share on other sites

yep thats the same one, around 60 or 70 pages, one interesting part was when they lined the port with a bunch of beads to see if it would perform like it would in theory. that paper helped me a lot in terms of grasping the basics of how to design and place a port, its always stuck in my mind

Fidelity

Built to Last

Team Sound Asleep

24Runner Build Log: http://www.stevemeadedesigns.com/board/topic/196657-24runner-sleeper-system-lots-of-fi-neo-dd-focal-new-video-w-juicebox-lithium/

2 x 12" Fi BTL N2 / 2 x 12" Fi BTL N3

2 DD M3b

Maxwell 2.7V 3000F Supercapacitors

Pioneer DEH-80prs

Focal P165 V30 components

Rockford Fosgate T-400.4

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

  • Recently Browsing   0 members

    • No registered users viewing this page.
  • Who's Online   1 Member, 0 Anonymous, 253 Guests (See full list)




×
×
  • Create New...